SEARCH

SEARCH BY CITATION

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results and discussion
  5. Experimental procedures
  6. Acknowledgements
  7. References.

We have visualized the relationship between the endoplasmic reticulum (ER) and Golgi in leaf cells ofNicotiana clevelandiiby expression of two Golgi proteins fused to green fluorescent protein (GFP). A fusion of thetrans-membrane domain (signal anchor sequence) of a rat sialyl transferase to GFP was targeted to the Golgi stacks. A second construct that expressed theArabidopsisH/KDEL receptor homologue aERD2, fused to GFP, was targeted to both the Golgi apparatus and ER, allowing the relationship between these two organelles to be studied in living cells for the first time. The Golgi stacks were shown to move rapidly and extensively along the polygonal cortical ER network of leaf epidermal cells, without departing from the ER tubules. Co-localization of F-actin in the GFP-expressing cells revealed an underlying actin cytoskeleton that matched precisely the architecture of the ER network, while treatment of cells with the inhibitors cytochalasin D and N-ethylmaleimide revealed the dependency of Golgi movement on actin cables. These observations suggest that the leaf Golgi complex functions as a motile system of actin-directed stacks whose function is to pick up products from a relatively stationary ER system. Also, we demonstrate for the first timein vivobrefeldin A-induced retrograde transport of Golgi membrane protein to the ER.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results and discussion
  5. Experimental procedures
  6. Acknowledgements
  7. References.

It is generally accepted that bidirectional transport of proteins between the endoplasmic reticulum (ER) and the Golgi in plant cells is by means of transport vesicles ( Denecke 1996;Staehelin & Moore 1995), although direct membrane connections have been described in some cell types ( Satiat-Jeunemaitre et al. 1996a ). The driving forces behind such transport and the molecular mechanisms underlying it are far from clear. However, in animal and yeast cells, the pathways of membrane and protein flow between the ER and Golgi have been dissected morphologically and biochemically ( Farquhar & Palade 1998). It is generally accepted that secretory product flow from the ER is by means of vesicle vectors transporting newly synthesized proteins and glycoproteins to an ‘intermediate’ compartment between the exit site on the ER and the cis-Golgi variously termed as vesicular tubular clusters (VTC), sorting exosomes or cis-Golgi compartments ( Aridor & Balch 1996). Here, the first sorting occurs for the return of proteins in the retrograde direction back to the ER. Subsequently, and depending on the model currently in vogue, this compartment either delivers its cargo to the cis-Golgi via COP-coated vesicles or matures into a cis-Golgi cisternum ( Bannykh et al. 1998 ).

Recent studies, utilizing a fusion of the green fluorescent protein (GFP) to a viral glycoprotein, have revealed the animal Golgi complex to be a relatively stationary structure, with rapidly forming pre-Golgi structures (VTC) being unidirectionally translocated by microtubules, inwards from the ER to the Golgi ( Presley et al. 1997 ). This new approach permits the in vivo observation of endomembrane dynamics and has been used previously to demonstrate the diffusional mobility of Golgi membrane proteins including GFP-tagged glycosyl transferases ( Cole et al. 1996 ).

We have demonstrated previously that GFP can be used for the in vivo study of the plant endomembrane system by targeting it to the ER lumen in tobacco leaf cells using a virus-based expression system ( Boevink et al. 1996 ). Here, in order to investigate the relationship between the leaf Golgi apparatus and ER we have expressed, in leaf cells, GFP spliced to the transmembrane domain (signal anchor sequence) of a rat sialyl transferase ( Munro 1995) and to the Arabidopsis homologue of the H/KDEL receptor ( Lee et al. 1993 ). We show that both these proteins target the tobacco leaf Golgi, which are highly motile and track in an actin-dependent manner over the polygonal network of cortical ER. This close relationship between the ER and Golgi is highlighted by a brefeldin A-induced retrograde transport of the sialyl transferase construct into the ER.

Results and discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results and discussion
  5. Experimental procedures
  6. Acknowledgements
  7. References.

Targeting GFP constructs to the Golgi

GFP was fused to the 52 N-terminal amino acids of a rat 2,6-sialyl transferase ( Munro 1995) encompassing the nine amino acid cytoplasmic domain, the trans-membrane domain (signal anchor sequence) and 26 amino acids of the luminal domain ( Fig. 1a,b). This chimeric protein (STtmd–GFP) was expressed in leaves of Nicotiana clevelandii following infection of plants with transcripts of the virus carrying the fusion protein gene. Three to four days after infection of the leaves, fluorescent infection sites were visible using a confocal laser scanning microscope (CLSM). In leaf epidermal cells, the fusion protein was observed in numerous, small fluorescent bodies in the cortical cytoplasm of the cells ( Fig. 2a), and also within trans-vacuolar strands of cytoplasm (data not shown). At higher magnifications these bodies were resolvable as doughnut-shaped or horseshoe-shaped structures that were in constant motion within the cytoplasm ( Fig. 2c, darts). To analyse this motion, time-lapse images were collected at 1.4-sec intervals and viewed as a movie sequence, which is accessible on our web site ( http://www.brookes.ac.uk/schools/bms/research/molcell/hawes/gfp/gfp.html). Immunogold labelling of tobacco leaf cells with an antibody to GFP demonstrated that the STtmd–GFP fusion protein was located in the Golgi bodies and appeared to be located predominantly in the trans-half of the stacks ( Fig. 3a).

image

Figure 1. (a) Schematic representation of the plasmids pTXS.STtmd–GFP and pTXS.ERD2–GFP (not drawn to scale) constructed from a truncated sialyl transferase gene ( Fig. 1b; Munro, 1995) and from the complete sequence of the Arabidopsis ERD2 homologue (Lee et al. 1993). In the viral vector, pTXS.P3C2, the viral cDNA is flanked by a T7 RNA polymerase promoter for production of transcripts and a SpeI restriction enzyme recognition site for linearization of the plasmid. The first four open reading frames of the viral genome are represented by the Mrs of their products (K = kDa) and the coat protein by CP. The duplicated coat protein subgenomic promoter is represented by black triangles and the multiple cloning site between the duplication is shown. The GFP fusions, represented below the vector, were inserted independently into the vector between the EagI and NsiI sites to give pTXS.STtmd–GFP and pTXS.ERD2–GFP.

(b) The complete coding sequence of sialyl transferase is represented diagrammatically with the position of the transmembrane domain (tmd) indicated by the black box. The 52 amino-terminal amino acids of sialyl transferase, indicated by the brackets, which were spliced onto GFP, include nine cytoplasmic amino acids at the amino terminus and the 17 amino acid transmembrane domain (boxed).

Download figure to PowerPoint

image

Figure 2. Expression of STtmd–GFP (a,c,h,i) and ERD2–GFP (b–g) in leaves of N. clevelandii.

(a) Confocal laser scanning micrograph of a PVX.STtmd–GFP-infected leaf epidermal cell showing the distribution of fluorescent Golgi in the cortical cytoplasm. The central, elliptical structure is an autofluorescent stomatal pore. Bar = 20 μm.

(b) Golgi stacks labelled with ERD2–GFP are closely associated with the polygonal network of cortical ER. Bar = 5 μm.

(c) High magnification views of individual living Golgi stacks reveal horseshoe- and doughnut-shaped configurations (darts) (cf. similar configuration in electron micrograph of an ERD2-labelled Golgi stack, Fig. 3b). Bar = 1 μm.

(d) Treatment of cells with cytochalasin D arrests the motion of the Golgi stacks and causes them to aggregate on small ‘islands’ of lamellar ER. Bar = 10 μm.

(e–g) Single epidermal cell revealing: (e) GFP-labelled ER/Golgi system; (f) actin cables labelled with rhodamine–phalloidin; (g) combination of images in (e) and (f) to show precise co-localization of Golgi stacks with underlying actin cables. Bar in (e) = 10 μm.

(h) Treatment of epidermal cells expressing STtmd–GFP (cf. (a) above) with BFA (50 μg ml–1) for 30 min causes the majority of Golgi bodies to disappear and the cortical ER to become fluorescent. A few Golgi bodies appear to be resistant to the drug and remain fluorescent. Bar = 10 μm.

(i) Washing the cells for 5 h on water induces the reformation of the Golgi stacks in the majority of epidermal cells, although some residual ER fluorescence persists. Bar = 20 μm.

Download figure to PowerPoint

image

Figure 3. Immunogold labelling of Golgi stacks with antibody to GFP.

(a) STtmd–GFP is located predominantly in the trans face of the Golgi stack. Bar = 100 nm.

(b) ERD2–GFP is distributed uniformly across the cisternae of the Golgi stack. Bar = 100 nm.

Download figure to PowerPoint

We next fused GFP to the carboxy terminus of the Arabidopsis thaliana homologue of the yeast HDEL receptor aERD2 ( Lee et al. 1993 ) and expressed this chimeric protein (ERD2–GFP) in N. clevelandii using the PVX expression vector ( Fig. 1a). In yeast and mammalian cells the ERD2 protein has been shown to be preferentially located at the cis-Golgi and the so-called ‘intermediate’ compartment, but cycles between these structures and the ER to retrieve escaped H/KDEL-containing ER resident proteins, back to the ER ( Townsley et al. 1993 ). Although the tissue specificity for the expression of the gene aERD2 in plants has been determined with higher levels in roots and leaf trichomes than in leaves, its subcellular location is unknown ( Bar-Peled et al. 1995 ). Expression of ERD2–GFP in PVX.ERD–GFP-infected N. clevelandii leaf epidermal cells also resulted in the production of fluorescent viral infection sites. Confocal microscopy of leaf segments revealed high levels of GFP expression in spherical bodies identical to those observed with STtmd–GFP, and also revealed a less fluorescent cortical network of tubular membranes ( Fig. 2b). This membranous network had the same morphology as the cortical ER described previously when an ER-targeted GFP–KDEL protein was expressed in tobacco leaves ( Boevink et al. 1996 ). Immunogold detection with the anti-GFP antibody again showed strong labelling of the Golgi stacks in fluorescing cells ( Fig. 3b), demonstrating for the first time the targeting of an H/KDEL receptor homologue to the Golgi apparatus of plant cells.

Motility of the Golgi apparatus

As already shown for the STtmd–GFP protein, Golgi stacks labelled with ERD2–GFP were highly motile in the cortical cytoplasm. However, as this construct also labelled the cortical ER, we were able to study the spatial and temporal relationships between the Golgi and the ER. Surprisingly, the Golgi stacks always remained closely associated with ER tubules, and individual stacks were not observed to occur away from the ER network, giving the Golgi the appearance of being tethered to the ER tubules (see the movie at http://www.brookes.ac.uk/schools/bms/research/molcell/hawes/gfp/gfp.html). Golgi movement was imaged using a multiple-frame averaging facility on the CLSM program. For instance, with eight 1-sec scans, multiple images of individual Golgi stacks moving along ER tubules could be obtained ( Fig. 4a). This trafficking of individual stacks along ER tubules could be measured from plots of time-lapse sequences made on individual epidermal cells ( Fig. 4b). Golgi stacks exhibited a range of movements as follows. (i) Directional tracking along the cortical ER. (ii) Tracking along the cortical ER and ‘jumping’ to a rapidly streaming strand of ER [such moving ER strands have been reported previously in Nicotiana epidermal cells ( Boevink et al. 1996 ) and also in onion bulb epidermis ( Lichtscheidl & Hepler 1996;Lichtscheidl & Url 1990)]. (iii) Stop/start movements. (iv) Stationary or saltatory movement. We measured rates of movement that varied from 0 to 0.76 μm sec–1 along the stationary cortical ER, and extremely rapid movement in excess of 2.2 μm sec–1 within trans-vacuolar cytoplasmic strands ( Fig. 4b).

image

Figure 4. (a) Image of the cortical ER obtained by eight sequential frame captures shows the movement of an individual Golgi stack (as multiple images) along a single ER strand. The direction of motion is indicated by the darts. Bar = 10 μm.

(b) Traces of the movement of five Golgi stacks along the cortical ER in a single epidermal cell demonstrate a variety of patterns of movement. (A) A slow stack progresses 8.9 μm in 102 sec, averaging 0.09 μm sec–1. (B) A Golgi stack migrating around the cortical ER network at 0.58 μm sec–1 subsequently ‘jumps’ onto a streaming cytoplasmic strand (between arrowheads) and rapidly accelerates to 2.2 μm sec–1. (C) A Golgi stack shows steady movement for 14 sec and then relatively slow movement over the next 54 sec. (D) A Golgi stack shows restricted, saltatory movement for 39 sec before accelerating to 0.15 μm sec–1. (E) A Golgi stack shows deceleration following 2 sec of rapid movement.

Download figure to PowerPoint

Video-enhanced microscopy of onion epidermal cells has suggested that the cortical ER can support the motility of a variety of unidentified organelles ( Lichtscheidl & Hepler 1996;Lichtscheidl & Url 1990) and it has been shown that the spatial organization of cortical ER is most likely under the control of the actin cytoskeleton not the cortical microtubules ( Lichtscheidl & Hepler 1996;Quader et al. 1989 ). In plants, the organization of the Golgi apparatus appears to be dependent on the actin cytoskeleton and not on microtubules ( Satiat-Jeunemaitre et al. 1996b ). To examine the spatial relationship between actin and the ER/Golgi system in leaf epidermal cells, we treated leaf segments with 20 μg ml–1 cytochalasin D to depolymerize the actin cytoskeleton. As expected, all movement of Golgi stacks along the ER network ceased within 40 min. After treatment, the fluorescent Golgi stacks were no longer associated with the tubular ER but with small islands of lamellar ER that occurred at the vertices of the cortical ER tubules ( Fig. 2d). Rhodamine–phalloidin staining of cytochalasin-treated epidermal cells confirmed the complete depolymerization of cytoplasmic actin filaments (data not shown). Thus, the overall polygonal organization of cortical ER was not dependent on actin, although cytochalasin treatment appeared to induce the formation of small islands of ER lamellae. Treatment with 0.5 m m N-ethylmaleimide, to arrest organelle movement dependent on cytoplasmic motors, completely inhibited all Golgi movement within 5 min of application. In this case, the stationary Golgi stacks remained associated with the tubular ER along which they were previously travelling (data not shown). The involvement of actin in Golgi movement was confirmed by the staining of cortical actin filaments with rhodamine-conjugated phalloidin. Mild fixation had only a minimal effect on the organization of the cortical ER and Golgi stacks. Beneath the ER network a corresponding network of actin cables was revealed. These actin cables were precisely overlaid by the ER tubules such that a vertex in the actin network was superimposed by a vertex in the ER network ( Fig. 2e,f). Electronic reduction in image contrast permitted the imaging of individual Golgi stacks and revealed them to be in close association with the actin filaments ( Fig. 2g). Thus, it appears that the cortical actin network forms a template on which the cortical ER is overlaid, or clad, and also functions as a ‘track’ along which the cortical Golgi stacks are translocated. To date we have no evidence on any factors controlling this precise alignment of actin filaments; however, it has been shown that microtubules can form similar patterns in epidermal cells ( Hoss & Wernick 1995) so their possible involvement in this system cannot be discounted.

Effects of brefeldin A on Golgi membranes

To investigate further the close spatial relationship between the cortical ER and Golgi stacks, segments of leaves expressing either STtmd–GFP or ERD2–GFP were incubated in the secretory inhibitor brefeldin A (BFA) at concentrations of 10 μg ml–1 and 50 μg ml–1. Both concentrations of BFA resulted in the disappearance of the majority of Golgi stacks from the epidermal cell cytoplasm. In cells expressing ERD2–GFP, the disappearance of the Golgi was accompanied by a corresponding increase in the brightness of GFP fluorescence associated with the ER (data not shown). In cells expressing STtmd–GFP, which showed no initial ER fluorescence ( Fig. 2a), 30 min of BFA treatment (50 μg ml–1) caused the majority of Golgi stacks to disappear and the cortical and trans-vacuolar ER to become fluorescent ( Fig. 2h). These data were corroborated by transmission electron microscopy of BFA-treated tobacco leaves, which showed a reduction in the number of Golgi cisternae with time following BFA treatment (data not shown). BFA did not inhibit the movement of those Golgi stacks that remained in the cell following treatment and did not affect general cytoplasmic streaming. These data provide the first evidence in plant cells for a retrograde transport of Golgi membrane protein into the ER in response to BFA, and support the evidence for a retrograde transport pathway as revealed by BFA treatment of mammalian cells ( Lippincott-Schwartz et al. 1990 ). This is in contrast to the reported BFA-induced vesiculation of Golgi in meristematic root tip cells, where there was no evidence of retrograde transport of Golgi material ( Satiat-Jeunemaitre & Hawes 1992a;Satiat-Jeunemaitre et al. 1996a ). As in previous reports of BFA action on plant cells, this response to the drug was completely reversible ( Satiat-Jeunemaitre & Hawes 1992a;Satiat-Jeunemaitre et al. 1996a ). Within 5 hours of floating leaf tissue on water, motile Golgi stacks could be observed in the majority of epidermal cells, although some residual fluorescence of the ER persisted ( Fig. 2i). To date, we have no evidence of the nature of the vectors involved in this retrograde transport from Golgi to ER. However, the distance of membrane transport must be minimal and the ER clearly has the capacity to reform Golgi stacks on, or in very close proximity to, its surface.

The data presented here have a number of important implications for our understanding of endomembrane dynamics in plant cells. We have demonstrated that plant cells possess (aERD2) and can support (STtmd) targeting mechanisms to the Golgi that are similar to those in yeast and mammalian cells ( Lee et al. 1993 ;Munro 1995, 1998), indicating some molecular homology between the yeast, animal and plant Golgi apparatus. Although there is no evidence for a native sialyl transferase in plants, the results presented here indicate a common mechanism for targeting and retention of transferases in Golgi membranes. It has been suggested that one mechanism by which Golgi glycan transferases are targeted to their correct location within the cisternal stacks is by a matching of the hydrophobic amino acids in the transmembrane domain to the width of the bilayer, which is determined by the sphingolipid and sterol composition of the membrane ( Munro 1995, 1998). However, an interaction between the cytoplasmic domain of the transferase and other native plant Golgi proteins cannot be excluded ( Munro 1998;Nilsson et al. 1993 ).

Our results also show that there is a close spatial relationship between the plant ER and Golgi stacks, a feature of living plant cells that has not previously been reported. In animal cells, the Golgi complex occupies a relatively stationary perinuclear location in the cell, with ER-to-Golgi transport occurring by unidirectional transport along microtubules ( Presley et al. 1997 ). In marked contrast, in leaf cells, the Golgi stacks are highly motile and are translocated on actin cables. The possibility exists for a novel form of membrane/protein transfer between the ER and the cis face of the plant Golgi in which the Golgi stack operates as a form of mobile ‘vacuum cleaner’, picking up products from the surface of the ER. Whether transfer between the two organelles occurs by transition vesicles or directly by short tubular connections remains to be demonstrated.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results and discussion
  5. Experimental procedures
  6. Acknowledgements
  7. References.

Construction of ERD2–GFP and STtmd–GFP

The coding sequence of the Arabidopsis ERD2 homologue was PCR amplified from the cDNA clone ( Lee et al. 1993 ). The sequence encoding the amino terminal 52 amino acids of the rat sialyl transferase was amplified from pSMH4 (kindly provided by S. Munro). These sequences were precisely fused to the amino terminus of the gfp gene using overlap extension PCR with appropriate oligonucleotides. The overlap extension PCR product was cloned into a potato virus X vector (pTXS.P3C2) ( Boevink et al. 1996 ) using EagI and NsiI restriction enzyme sites engineered into the flanking oligonucleotides. Standard DNA manipulation techniques were used ( Sambrook et al. 1989 ). The resultant plasmids, pTXS.ERD2–GFP and pTXS.STtmd–GFP, carried the chimeric gfp genes under the transcriptional control of duplicated subgenomic RNA promoters ( Fig. 1).

In vitro transcription and plant inoculation

Run-off transcripts were synthesized from SpeI linearized pTXS.ERD2–GFP and pTXS.STtmd–GFP DNA using a T7 transcription kit (Ambion). Approximately 4-week-old N. clevelandii plants were inoculated by rubbing aluminium oxide-dusted leaves with transcripts derived from 0.2 mg of plasmid template.

Fluorescence and electron microscopy

Cells infected by the modified viruses expressing the chimeric GFP were imaged with a Bio-Rad MRC 1000 confocal laser scanning microscope (CSLM) as previously described by ( Oparka et al. 1995 ), or a Zeiss LSM 410 confocal laser scanning microscope. For electron microscopy, segments of fluorescing leaf tissue were low-temperature embedded in LR White resin for electron microscopy, sectioned and immunogold labelled with an anti-GFP polyclonal antibody (Molecular Probes) and 10 nm gold-conjugated secondary antibody, essentially using the protocol of Satiat-Jeunemaitre & Hawes (1992a). Micrographs were taken with a JEOL 1200EXII electron microscope.

To stain the actin cytoskeleton, pieces of virus-infected leaves containing ERD2–GFP and STtmd–GFP were excised and attached to 1 × 1 cm plastic supports. The lower epidermis and mesophyll was then removed by abrasion with emery cloth. The remaining tissue (upper epidermis and palisade layer) was pre-incubated for 30 min with 0.9 m m 3-maleido-benzoyl N-hydroxysuccinimide ester in buffer to stabilize the actin ( Sonobe & Shibaoka 1989) and then fixed for 3 min in 3% glutaraldehyde. The fixed tissue was stained in a 0.1 m m solution of rhodamine-conjugated phalloidin (Molecular Probes) for 30 min, excised from the plastic supports, and observed with the CLSM.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Results and discussion
  5. Experimental procedures
  6. Acknowledgements
  7. References.

We thank Sean Munro (MRC Laboratory for Cell Biology, Cambridge) for the kind gift of the rat sialyl transferase gene, and Natasha Raikhel (MSU-DOE Plant Research Laboratory, Michigan State University) for aERD2. This work was supported by a Leverhulme Trust grant.

References.

  1. Top of page
  2. Summary
  3. Introduction
  4. Results and discussion
  5. Experimental procedures
  6. Acknowledgements
  7. References.
  • Aridor, M. & Balch, W.E. 1996 Principles of selective transport: coat complexes hold the key. Trends Cell Biol., 6, 315 320.
  • Bannykh, S.I., Nishimura, N., Balch, W.E. 1998 Getting into the Golgi. Trends Cell Biol., 8, 21 25.
  • Bar-Peled, M., Da Silva Conceiçïo, A., Frigerio, L., Raikhel, N.V. 1995 Expression and regulation of aERD2, a gene encoding the KDEL receptor homolog in plants, and other genes encoding proteins involved in ER–Golgi vesicular trafficking. Plant Cell, 7, 667 676.
  • Boevink, P., Santa Cruz, S., Hawes, C., Harris, N., Oparka, K.J. 1996 Virus-mediated delivery of the green fluorescent protein to the endoplasmic reticulum of plant cells. Plant J., 10, 935 941.
  • Cole, N.B., Smith, C.L., Sciaky, N., Terasaki, M., Edidin, M., Lippincott-Schwartz, J. 1996 Diffusional mobility of Golgi proteins in membranes of living cells. Science, 273, 797 801.
  • Denecke, J. 1996 Mechanisms controlling function, identity and integrity of the plant endoplasmic reticulum. In: Membranes: Specialised Functions in Plants ( Smallwood, M. Knox, J.P. & Bowls, D.J., eds.). Oxford: Bios Scientific Publishers, pp. 367 382.
  • Farquhar, M.G. & Palade, G.E. 1998 The Golgi apparatus: 100 years of progress and controversy. Trends Cell Biol., 8, 2 10.
  • Hoss, S. & Wernicke, W. 1995 Microtubules and the establishment of apparent cell wall invaginations in mesophyll cells or Pinus sylvestris. J. Plant Physiol., 147, 474 476.
  • Lee, H.-I., Gal, S., Newman, T.C., Raikhel, N.V. 1993 The Arabidopsis endoplasmic reticulum retention receptor functions in yeast. Proceedings Natl Acad. Sci. USA, 90, 11433 11437.
  • Lichtscheidl, I.K. & Hepler, P.K. 1996 Endoplasmic reticulum in the cortex of plant cells. In: Membranes: Specialised Functions in Plants ( Smallwood, M. Knox, J.P. & Bowls, D.J., eds.). Oxford: Bios Scientific Publishers, pp. 383 402.
  • Lichtscheidl, I.K. & Url, W.G. 1990 Organisation and dynamics of cortical endoplasmic reticulum in inner epidermal cells of onion bulb scales. Protoplasma, 157, 203 215.
  • Lippincott-Schwartz, J., Donaldson, J., Schweizer, A., Berger, E., Hauri, H., Yuan, L., Klausner, R. 1990 Microtubule-dependent retrograde transport of proteins into the ER in the presence of Brefeldin A suggests an ER recycling pathway. Cell, 60, 821 836.
  • Munro, S. 1995 An investigation into the role of trans-membrane domains in Golgi protein retention. EMBO J., 14, 4695 4704.
  • Munro, S. 1998 Localization of proteins to the Golgi apparatus. Trends Cell Biol., 8, 11 15.
  • Nilsson, T., Slusarewicz, P., Hoe, M.H., Warren, G. 1993 Kin recognition: a model for the retention of Golgi enzymes. FEBS Lett., 330, 1 4.
  • Oparka, K.J., Roberts, A.G., Prior, D.A.M., Chapman, S., Baulcombe, D., Santa Cruz, S. 1995 Imaging the green fluorescent protein in plants – viruses carry the torch. Protoplasma, 189, 133 141.
  • Presley, J.F., Cole, N.B., Schroer, T.A., Hirschberg, K., Zaal, K.J.M., Lippincott-Schwartz, J. 1997 ER-to-Golgi transport visualized in living cells. Nature, 389, 81 85.
  • Quader, H., Hofmann, A., Schnepf, E. 1989 Reorganization of the endoplasmic reticulum in epidermal cells of onion bulb scales after cold stress: involvement of cytoskeletal elements. Planta, 177, 273 280.
  • Sambrook, J., Fritsch, E.F., Maniatis, T. 1989 Molecular Cloning. A Laboratory Manual, 2nd Edn. Cold Spring Harbor, NY: Cold Spring, Harbor Laboratory Press.
  • Satiat-Jeunemaitre, B., Cole, L., Bourett, T., Howard, C., Hawes, C. 1996a Brefeldin A effects in plant and fungal cells: something new about vesicle trafficking? J. Microsc., 181, 162 177.
  • Satiat-Jeunemaitre, B. & Hawes, C. 1992a Redistribution of a Golgi glycoprotein in plant cells treated with Brefeldin A. J. Cell Sci., 103, 1153 1166.
  • Satiat-Jeunemaitre, B., Steel, C., Hawes, C. 1996b Golgi-membrane dynamics are cytoskeleton dependent. Protoplasma, 191, 21 23.
  • Sonobe, S. & Shibaoka, H. 1989 Cortical fine actin filaments in higher plant cells visualized by rhodamine-phalloidin after pretreatment with m-maleimidobenzoyl-N-hydroxysuccinimide ester. Protoplasma, 148, 80 86.
  • Staehelin, L.A. & Moore, I. 1995 The plant Golgi apparatus: structure, functional organization and trafficking mechanisms. Annu. Rev. Plant Physiol. Plant Mol. Biol., 46, 261 288.
  • Townsley, F.M., Wilson, D.W., Pelham, H.R.B. 1993 Mutational analysis of the human KDEL receptor: distinct structural requirements for Golgi retention, ligand binding and retrograde transport. EMBO J., 12, 1557 1574.